检索结果-内蒙古大学图书馆

Sign language recognition using the fusion of image and hand landmarks through multi-headed convolutional neural network

引用

Scientific reports 2023年第1期13卷 16975页

作者： Refat Khan Pathan Munmun Biswas Suraiya Yasmin Mayeen Uddin Khandaker Mohammad Salman Ahmed A F Youssef Department of Computing and Information Systems School of Engineering and Technology Sunway University 47500 Bandar Sunway Selangor Malaysia. Department of Computer Science and Engineering BGC Trust University Bangladesh Chittagong 4381 Bangladesh. Department of Computer and Information Science Graduate School of Engineering Tokyo University of Agriculture and Technology Koganei Tokyo 184-0012 Japan. Centre for Applied Physics and Radiation Technologies School of Engineering and Technology Sunway University 47500 Bandar Sunway Selangor Malaysia. mayeenk@sunway.edu.my. Faculty of Graduate Studies Daffodil International University Daffodil Smart City Birulia Savar Dhaka 1216 Bangladesh. mayeenk@sunway.edu.my. College of Engineering and Technology American University of the Middle East Egaila Kuwait.

Sign Language Recognition is a breakthrough for communication among deaf-mute society and has been a critical research topic for years. Although some of the previous studies have successfully recognized sign language, it requires many costly instruments including sensors, devices, and high-end processing power. However, such drawbacks can be easily overcome by employing artificial intelligence-based techniques. Since, in this modern era of advanced mobile technology, using a camera to take video or images is much easier, this study demonstrates a cost-effective technique to detect American Sign Language (ASL) using an image dataset. Here, "Finger Spelling, A" dataset has been used, with 24 letters (except j and z as they contain motion). The main reason for using this dataset is that these images have a complex background with different environments and scene colors. Two layers of image processing have been used: in the first layer, images are processed as a whole for training, and in the second layer, the hand landmarks are extracted. A multi-headed convolutional neural network (CNN) model has been proposed and tested with 30% of the dataset to train these two layers. To avoid the overfitting problem, data augmentation and dynamic learning rate reduction have been used. With the proposed model, 98.981% test accuracy has been achieved. It is expected that this study may help to develop an efficient human-machine communication system for a deaf-mute society.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Graphic Environment for Generating Automated Tests and Documentation for ADAS

Graphic Environment for Generating Automated Tests and Docum...

引用

Zooming Innovation in Consumer Electronics International Conference (ZINC)

作者： Dominik Birtić Mario Vranješ Željko Lukač Gordana Velikić Institute RT-RK Osijek LLC for Information Technology Cara Hadrijana 10b Osijek Croatia Faculty of Electrical Engineering Computer Science and Information Technology Osijek Kneza Trpimira 2B Osijek Croatia Faculty of Technical Sciences Novi Sad Trg Dositeja Obradovića 6 Novi Sad Serbia RT-RK Institute for Computer Based Systems Narodnog Fronta 23a Novi Sad Serbia

ISBN: (数字)9781728182599

ISBN: (纸本)9781728182605

Advanced driver-assistance systems (ADAS) are primarily intended to help drivers in traffic and to increase driving safety. Today a large number of engineers are developing different algorithms for ADAS. This results in large quantity of written program codes on a daily basis, which has to be tested. Note that a manual testing and writing a code for the manual testing is an arduous task. To satisfy the increasing testing needs, and to accelerate the testing process, we developed a graphical environment that allows users to create automated tests quickly and efficiently by a simple drag and drop method. The environment was created using a *** server, MongoDB and Blockly. Blockly was used to create blocks and to combine blocks to make scripts, i.e., automated tests for ADAS. *** was used to save blocks into the MongoDB database and to load existing blocks into the environment. The environment enables users to use existing blocks to create Python scripts and their own blocks which can be used to create Python scripts, and to edit or remove existing blocks. The environment was tested manually. Testing results showed that all functionalities work properly and that the environment enables users to generate different scripts for automated testing in an order of seconds.

关键词： Testing Servers Documentation Databases Python Production facilities

来源：评论

学校读者我要写书评

暂无评论

Modification and optimization of solovey-strassen's fast exponentiation probablistic test binary algorithm

Modification and optimization of solovey-strassen's fast exp...

引用

2019 IEEE East-West Design and Test Symposium, EWDTS 2019

作者： Myzdrikov, Nikita Ye Semeonov, Ivan Ye. Yukhnov, Vasiliy I. Safaryan, Olga A. Reshetnikova, Irina V. Lobodenko, Andrey G. Cherckesova, Larissa V. Porksheyan, Vitaliy M. Information System Cybersecurity Don State Technical University Rostov-on-Don Russia Applied Mathematics Don State Technical University Rostov-on-Don Russia Info Communicative Technologies and Communication Systems Moscow Technical University of Communication and Computer Science Rostov-on-Don Russia Communication Systems in the Railway Transportation Rostov State University of the Means of Communication Rostov-on-Don Russia Information Systems and Radioengineering Don State Technical University Rostov-on-Don Russia Mathematics and Computer Sciences Cair Don State Technical University Rostov-on-Don Russia

This article will consider the probability test of Solovey-Strassen, to determine the simplicity of the number and its possible modifications. This test allows for the shortest possible time to determine whether the n... 详细信息

ISBN: (纸本)9781728110035

关键词： Jacobian matrices C# languages Cryptography Probabilistic logic Software algorithms information systems

来源：评论

学校读者我要写书评

暂无评论

Multi-channel Convolutions Neural Network Based Diabetic Retinopathy Detection from Fundus Images 16

Multi-channel Convolutions Neural Network Based Diabetic Ret...

引用

16th International Learning and Technology Conference, L and T 2019

作者： Butt, M. Mohsin Latif, Ghazanfar Iskandar, D.N.F. Awang Alghazo, Jaafar Khan, Adil H. Faculty of Computer Science and Information Technology Universiti Malaysia Sarawak Malaysia College of Computer Engineering and Sciences Prince Mohammad Bin Fahd University Khobar Saudi Arabia College of Applied and Supporting Studies King Fahd University of Petroleum and Minerals Saudi Arabia College of Engineering Prince Mohammad Bin Fahd University Saudi Arabia

Diabetic Retinopathy (DR) is an eye medical condition usually found in patients suffering from diabetes. The initial stages of DR can either show no symptoms or cause mild vision problems but advanced stages of the disease can cause blindness. The main cause of DR is high blood sugar in diabetic patients that affect the blood vessels supplying blood to the retina and blocks them. The body tries to grow new vessels to supply the retina, however, they usually don't develop properly and can easily leak. DR Detection is an extremely challenging task due to the variation of retina change throughout the disease stages. In this paper, a multi- channel Convolutional Neural Network (CNN) is proposed for DR detection from fundus images of the eyes. The proposed system is tested on a DR Dataset consisting of 35,126 images provided by EyePACS. Experimental results indicate that the accuracy of 97.08% is achieved through the model that outperforms those achieved through other methods in recent studies. © 2019 The Authors. Published by Elsevier B.V.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

Preface

EAI/Springer Innovations in Communication and Computing

引用

EAI/Springer Innovations in Communication and Computing 2022年 v-vii页

作者： Gao, Honghao Kim, Jung Yoon Hussain, Walayat Iqbal, Muddesar Duan, Yucong Shanghai University Room D319 Shanghai China Gachon University Seongnam-si Korea Republic of School of Information Systems and Modelling Faculty of Engineering IT University of Technology Sydney UltimoNSW Australia Computer Science and Informatics Division School of Engineering London South Bank University London United Kingdom Computer Science Hainan University Haikou Hainan China

来源：评论

学校读者我要写书评

暂无评论

Comparison of power factor between permanent magnet assisted synchronous reluctance motor with neodymium-iron-boron and ferrite magnets

Comparison of power factor between permanent magnet assisted...

引用

International Symposium on Power Electronics, Electrical Drives, Automation and Motion, SPEEDAM

作者： Alireza Siadatan Jacob Goltieb Ebrahim Afjei Energy Systems Group Faculty of Applied Science and Engineering University of Toronto Toronto ON Canada Department Electrical & Computer Eng. University of British Columbia Vancouver BC Canada Department of Electrical Eng. Shahid Beheshti University G.C. Tehran IRAN

ISBN: (数字)9781728170190

ISBN: (纸本)9781728170206

In this paper, the design of a permanent magnet assisted synchronous reluctance motor (PMASynRM or PMASRM) is presented as a replacement for a suitable permanent magnet in a simulated synchronous reluctance motor (synrm). In addition to solving the important problems faced by typical synchronous reluctance motors (low power factor and high torque ripple) it can provide more power of coping and then other common motors such as induction motors (IM). A common motor model was created for simulation in Maxwell finite element software. Then, ferrite and neodymium-iron-boron (NdFeB) magnets were compared as the permanent magnet The results show that the fields of average torque, Back-emf voltage and efficiency, and the torque ripple are all greater in the neodymium-iron-boron magnet than the ferrite magnet. Finally the power factors for both magnets are calculated from the power, voltage and current simulations as 0.68 in ferrite magnet and 0.89 in neodymium-iron-boron magnet. The results also indicated that a higher power factor in the second model than the first one.

关键词： Magnetic flux Reluctance motors Ferrites Magnetic separation Torque Induction motors

来源：评论

学校读者我要写书评

暂无评论

Regional Image Perturbation Reduces Lp Norms of Adversarial Examples While Maintaining Model-to-model Transferability

arXiv

引用

arXiv 2020年

作者： Ozbulak, Utku Peck, Jonathan de Neve, Wesley Goossens, Bart Saeys, Yvan van Messem, Arnout Department of Electronics and Information Systems Ghent University Ghent Belgium Center for Biotech Data Science Ghent University Global Campus Incheon Korea Republic of Department of Applied Mathematics Computer Science and Statistics Ghent University Ghent Belgium Data Mining and Modeling for Biomedicine VIB Inflammation Research Center Ghent Belgium Department of Telecommunications and Information Processing Ghent University - imec Ghent Belgium

Regional adversarial attacks often rely on complicated methods for generating adversarial perturbations, making it hard to compare their efficacy against well-known attacks. In this study, we show that effective regional perturbations can be generated without resorting to complex methods. We develop a very simple regional adversarial perturbation attack method using cross-entropy sign, one of the most commonly used losses in adversarial machine learning. Our experiments on ImageNet with multiple models reveal that, on average, 76% of the generated adversarial examples maintain model-to-model transferability when the perturbation is applied to local image regions. Depending on the selected region, these localized adversarial examples require significantly less Lp norm distortion (for p ∈ {0, 2, ∞}) compared to their non-local counterparts. These localized attacks therefore have the potential to undermine defenses that claim robustness under the aforementioned norms. Copyright © 2020, The Authors. All rights reserved.

关键词： Perturbation techniques

来源：评论

学校读者我要写书评

暂无评论

Deep Learning in Cancer Diagnosis and Prognosis Prediction: A Minireview on Challenges, Recent Trends, and Future Directions

引用

Computational and Mathematical Methods in Medicine 2021年第1期2021卷 9025470页

作者： Tufail, Ahsan Bin Ma, Yong-Kui Kaabar, Mohammed K. A. Martínez, Francisco Junejo, A.R. Ullah, Inam Khan, Rahim School of Electronics and Information Engineering Harbin Institute of Technology Harbin150001 China Department of Electrical and Computer Engineering COMSATS University Islamabad Sahiwal Campus Sahiwal57000 Pakistan Gofa Camp Near Gofa Industrial College and German Adebabay Nifas Silk-Lafto Addis Ababa26649 Ethiopia Institute of Mathematical Sciences Faculty of Science University of Malaya Kuala Lumpur50603 Malaysia Department of Applied Mathematics and Statistics Technological University of Cartagena Cartagena30203 Spain School of Control Science and Control Engineering Harbin Institute of Technology Harbin150001 China Changzhou Campus 213022 China

Deep learning (DL) is a branch of machine learning and artificial intelligence that has been applied to many areas in different domains such as health care and drug design. Cancer prognosis estimates the ultimate fate of a cancer subject and provides survival estimation of the subjects. An accurate and timely diagnostic and prognostic decision will greatly benefit cancer subjects. DL has emerged as a technology of choice due to the availability of high computational resources. The main components in a standard computer-aided design (CAD) system are preprocessing, feature recognition, extraction and selection, categorization, and performance assessment. Reduction of costs associated with sequencing systems offers a myriad of opportunities for building precise models for cancer diagnosis and prognosis prediction. In this survey, we provided a summary of current works where DL has helped to determine the best models for the cancer diagnosis and prognosis prediction tasks. DL is a generic model requiring minimal data manipulations and achieves better results while working with enormous volumes of data. Aims are to scrutinize the influence of DL systems using histopathology images, present a summary of state-of-the-art DL methods, and give directions to future researchers to refine the existing methods. © 2021 Ahsan Bin Tufail et al.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Approximation of distortion sound via Fourier and wavelet transform 25

Approximation of distortion sound via Fourier and wavelet tr...

引用

25th International Congress on Sound and Vibration 2018: Hiroshima Calling, ICSV 2018

作者： Suzuki, Toshio Zempo, Keiichi Kinoshita, Tamotu Ryutsu Keizai University Faculty of Distribution and Logistics Systems Japan University of Tsukuba Faculty of Engineering Systems and Information Japan University of Tsukuba Faculty of Pure and Applied Science Japan

ISBN: (纸本)9781510868458

Music dictation is a very popular way to play music without using musical scores. However, this skill depends on one’s experience and sense. Moreover, the effects units such as distortion, chorus, flanger and so on control the tones of guitar sounds used in recent music. This makes it difficult to take down in musical notation for not only human also software. In this paper, we focus on the effect of distortion that is one of the most famous guitar effect units. Our motivation is to improve music dictation regardless of one’s skills and experience in the sense of not the pitch but the quality of the sound. Distortion units involve a nonlinear transform, which is constructed by the following two steps: amplification and clipping. It is difficult to analyze distortion sounds by Fourier method, since the distortion unit process has clipping;the nonlinear transform. We have defined the methods to extract the feature of distortion sound with the wavelet. That is, we defined the index of distortion sound feature as the correlation between the sound signal and the Haar wavelet. Since the wavelet function has the strong localization, we can extract the feature of distortion sound. Therefore, we propose the describing method based on the index of distortion written above and the center of gravity of the spectra, which corresponds to the tone of the sound. The subjective experiments were conducted to validate the similarity between the proposed indexing method and the human perception. The result of the experiment showed that the sounds with same feature value and same center of gravity of the spectrum are recognized as the similar sounds. This means the proposed method could be used as the index of the distortion sounds which human and machine could commonly recognize. © 25th International Congress on Sound and Vibration 2018, ICSV 2018: Hiroshima Calling. All rights reserved.

关键词： Wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

IoT-based Smart Parking System using Android Application

IoT-based Smart Parking System using Android Application

引用

Computing, Communication and Electronics (ETCCE), Emerging Technology in

作者： Nor Bakiah Abd Warif Mohd Izzat Syahmi Saiful Azman Nor-Syahidatul N Ismail Muhammad Akmal Remli Faculty of Computer Science & Information Technology Universiti Tun Hussien Onn Malaysia Batu Pahat Johor Malaysia Faculty of Computing College of Computing and Applied Sciences Universiti Malaysia Pahang Pekan Pahang Malaysia Institute For Artificial Intelligence and Big Data Universiti Malaysia Kelantan (Kota Campus) Kota Bharu Kelantan Malaysia

ISBN: (数字)9781665419628

ISBN: (纸本)9781665419635

Since all people used vehicles as a necessity, the growing number of vehicles is increasing day by day. With the increasing number of vehicles on the road, the requirement of parking lot will also be restricted. Thus, it become a problem for the people to find a free parking space in the busy building especially in shopping complex. Internet of Things (IoT) is an interesting technology that can be applied in this paper because this technology has a capability to transfer the data over a network without requiring human intervention in almost all applications in today's society. In this paper, a user-friendly mobile application, named as Android-based Car Parking Monitoring System (ACPMS) is built to aid in locating a particular parking place. ACPMS can provide a user with the ability to check vacant parking space and locate the nearest parking lot. ACPMS obtained the parking location from the current user's position with the sensor located in the shopping complex's parking lot. ACPMS is testing in a realistic environment for the purpose of movement detection and location service to notify user using the mobile application. By considering seven test case scenarios, the combination of the ACPMS mobile application with a parking prototype kit shows that the proposed work able to solve the parking problem.

关键词： Databases Wires Prototypes Hardware Mobile applications Internet of Things Smart phones

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：